Polyethylenimine Increases Antibacterial Efficiency of Chlorophyllin

Polyethylenimines (PEIs), a group of polycationic molecules, are known to impair the outer membrane of Gram-negative bacteria and exhibit antimicrobial activity. The outer membrane of Gram-negative strains hinders the uptake of photosensitizer chlorophyllin. In this study, we report chlorophyllin and branched PEI combinations’ activity against Escherichia coli strains DH5α and RB791, Salmonella enterica sv. Typhimurium LT2, and Bacillus subtilis 168. The minimal bactericidal concentration (MBC) was determined by plating cells treated with different concentrations of PEI and chlorophyllin on agar and monitoring their growth after 24 h. All tested combinations of PEI and chlorophyllin were lethal for S. enterica after 240 min of incubation in light, whereas PEI alone (<100 µg mL−1) was ineffective. In the darkness, complete inhibition was noted with a combination of ≥2.5 µg mL−1 chlorophyllin and 50 µg mL−1 PEI. If applied alone, PEI alone of ≥800 µg mL−1 of PEI was required to completely inactivate E. coli DH5α cells in light, whereas with ≥5 µg mL−1 chlorophyllin, only ≥100 µg mL−1 PEI was needed. No effect was detected in darkness with PEI alone. However, 1600 µg mL−1 PEI in combination with 2.5 µg mL−1 resulted in complete inactivation after 4 h dark incubation. PEI alone did not inhibit E. coli strain RB791, while cells were inactivated when treated with 10 µg mL−1 chlorophyllin in combination with ≥100 µg mL−1 (in light) or ≥800 µg mL−1 PEI (in darkness). Under illumination, B. subtilis was inactivated at all tested concentrations. In the darkness, 1 µg mL−1 chlorophyllin and 12.5 µg mL−1 PEI were lethal for B. subtilis. Overall, PEI can be used as an antimicrobial agent or potentiating agent for ameliorating the antimicrobial activity of chlorophyllin.


Introduction
The discovery of antibiotics in the last century and their subsequent application against pathogens resulted in a significant reduction in infectious diseases and an increase in the average life expectancy [1]. However, the overuse and misuse led to the development of resistance against almost all available antibiotics [2]. Today, antibiotic resistance is a significant threat to public health, and if no countermeasures are taken, it may cause

Effect of PEI and Chlorophyllin on Bacillus subtilis 168
The Gram-positive model bacterium B. subtilis was inactivated considerably quickly in the presence of chlorophyllin and PEI combinations (Figure 3). The sensitivity of B. subtilis against PEI and chlorophyllin was much higher in light conditions. Complete inhibition of cells was observed at 0 min of analysis (corresponds to 15 min of sample preparation and sampling in low light conditions) and PEI concentrations of 100 µg mL −1 and above in light incubated cells. At 0 min, chlorophyllin alone did not exhibit an immediate effect. However, 60 min sample analysis showed that ≥0.004 µg mL −1 of chlorophyllin alone completely eradicated the cells in light conditions (Figure 3). Light incubated cells were already entirely eradicated at the lowest tested concentrations of chlorophyllin (0.002 µg mL −1 ) and PEI (>12.5 µg mL −1 ). The cells were killed within 120 min of incubation in light at all the tested concentrations of PEI (12.5-800 µg mL −1 ) and chlorophyllin (0.002-2.000 µg mL −1 ) combinations. Only control cells (no chlorophyllin and PEI) showed growth. Pronounced synergistic effects were detected above all at 0 min. Due to the strong effect of both components on bacteria survival, Combenefit analysis can hardly detect synergism at later time points. Complete (100%) inhibition of cells was observed at 2 µg mL −1 of chlorophyllin alone and >12.5 µg mL −1 of PEI alone at 120 min of incubation in darkness ( Figure 3).

Determination of the Effect of PEI and Chlorophyllin Related to Cell Density
All the above experiments were performed with a defined culture density (OD 600nm ) of 0.01. The MBC of different concentrations of PEI (1.95-250 µg mL −1 ) + chlorophyllin (0-30 µg mL −1 ) was investigated against different initial cell densities (OD 600nm = 0.0003-0.73). For this study, freshly grown E. coli DH5α was employed, and the experiment was performed in 96-well microtiter plates. In light, higher concentrations of PEI-chlorophyllin showed stronger eradication effects on higher cell concentrations. Cell densities ≥ 0.0007 (OD 600nm ) incubated in light conditions were completely eradicated at the highest tested concentration of PEI (250 µg mL −1 ) ( Table 2). Chlorophyllin alone was not effective in both light and dark incubation conditions. However, in the presence of PEI (≥125 µg mL −1 ) and under illumination, a strong effect was noticed at 30 µg mL −1 chlorophyllin for the tested highest cell density (i.e., OD 600nm = 0.73). Dark controls, in contrast, were hardly affected at the applied concentration range.    The maximum cell densities that are entirely eradicated at a given combination of PEI (vertical concentration gradient in µg mL −1 ) and chlorophyllin (horizontal concentration gradient in µg mL −1 ) are presented (n.e.: no effect).
x ≈ √ t × 2D, where x is the distance, t is the lifetime, and D is the diffusion coefficient of 1 × 10 −5 cm 2 [35]). Consequently, only superoxides (e.g., H 2 O 2 ) can affect cellular targets at a longer distance. This led to the assumption that chlorophyllin present outside the cell is less effective in producing cell damage than chlorophyllin accumulated in the cytoplasm. The outer membrane of Gram-negative bacteria was found to be a barrier for chlorophyllin, significantly reducing its efficiency [9,10]. The treatment of Gram-negative bacteria with a combination of chlorophyllin and a cell wall destabilizing agent colistin led to the uptake of chlorophyllin into the cell cytoplasm and increased cytotoxicity, leading to photodynamic cell death.
Our data show that branched PEI has significant potential as synergistic or potentiating agents against bacteria. However, in the course of experiments using the cell wall destabilizing agent colistin in combination with chlorophyllin, the necessary concentration (weight per volume) of colistin for bacterial eradication was found to be far below that required by branched PEI. E. coli DH5α was eradicated after 4 h of incubation with concentrations of 0.5 µg/mL colistin and 5 µg/mL chlorophyllin compared to 100 µg/mL PEI in combination with the same concentration of chlorophyllin. The difference is probably due to the different structures of the molecules. Colistin is smaller (1155.4 g/mol) compared to the branched PEI (about 25 kDa) employed in this study. In addition, the colistin molecule contains hydrophilic as well as lipophilic components, while the polybasic amine polymer branched PEI (B-PEI) is water soluble and lacks lipophilic components [10,36]. PEI is available in various forms (linear, branched, and dendrimer) and sizes (800-270,000 kDa, Sigma-Aldrich, S. Louis, MO, U.S.A.), which were not yet tested for their eradication potential in combination with chlorophyllin. Additionally, chemical modifications have the potential for increasing the effectivity of PEI. Yin et al. (2017) synthesized various quaternized forms of PEI800 by nucleophilic substitution with different alkyl bromide chains [37]. They tested antibacterial effects against the Gram-positive strain MRSA70069 and the Gram-negative Pseudomonas aerguinosa 902, as well as membrane depolarization (using the membrane potential-sensitive dye diSC3 (5)). All quaternized PEI800-polymers exhibited increased cytotoxicity against bacteria and induced significant membrane depolarization compared to unmodified PEI800. PEI800 FQ C6, a PEI800 fully quaternized with short alkyl groups, was the most effective PEI polymer. In another study, Roest et al. (2015) examined the charging properties of hyperbranched polyurea-polyethyleneimine coatings [25]. After varying the degree of alkylation, the authors concluded that cationic charges alone are ineffective in contact-killing bacteria and needs to be carried by an alkylated nitrogen species. To investigate the effect of PEI on bacteria causing healthcare-associated infections, Azevedo et al. (2014) studied the effect of PEI and PEI-based nanoparticles (synthesized by reductive amination) on various bacterial species (Staphylococcus aureus, Staphylococcus epidermidis, Acinetobacter baumannii) and the yeast Candida albicans as well as their ability to form biofilms in a polyurethane catheter-model system [38]. PEI, in particular, inhibited growth and biofilm formation very efficiently, while nanoparticles were found to be less effective in these experiments. In contrast, another study using quaternary ammonium polyethyleneimine-(QA-PEI) nanoparticles resulted in the effective inactivation of S. aureus and E. coli [39]. The addition of PEI-nanoparticles into resin composite materials showed long-lasting antibacterial and anti-biofilm effects, making PEI interesting for dental prosthetics [40]. Following promising ex vivo experiments, application in the antimicrobial treatment of dental root canals [41] and bone cement for joint replacement surgery was suggested [40].
The Gram-positive model bacterium B. subtilis was severely affected by combinations of chlorophyllin and PEI. Under illumination, a complete loss of viability was found after 60 min at all tested concentrations of PEI, chlorophyllin, or their combinations, respectively. Interestingly, the highest tested concentration of chlorophyllin (2 µg/mL) alone did not completely eradicate the cells in darkness; however, combination with 12.5 µg/mL PEI led to complete eradication of the cells. This result comes in agreement with Foxley et al., (2016), who found that susceptibility of methicillin-resistant S. aureus (MRSA) strains against ampicillin and vancomycin can be restored in the presence of branched PEI because of its interaction with the wall teichoic acids (WTA) without entering the cells [42]. Generally, Gram-positive bacteria synthesize their cell wall with certain transpeptidases (PBPs, penicillin-binding proteins), which are the target of β-lactams. MRSA, however, possesses a β-lactam-PBP2a, providing resistance against this group of antibiotics. Positively charged amino groups of the branched PEI interact with negatively charged phosphate bridges of WTAs and impede the appropriate orientation of PBP2a, a prerequisite of cell wall synthesis that requires WTAs for its proper activity, thus, implying that the activity of PEI is most likely due to electrostatic interaction [43].
From the above-mentioned data, we can conclude that PEI can be used as an antimicrobial agent or potentiating agent for ameliorating the antimicrobial activity of chlorophyllin or other antibiotics.

Bacterial Strains and Growth Conditions
Experiments were performed with bacterial strains E. coli DH5α, E. coli RB791, S. Typhimurium LT2, and B. subtilis 168. Cells were grown overnight (16 ± 2 h) in Luria Bertani (LB) medium at 37 • C. Cell concentration was determined by measuring the culture's optical density (OD) at 600 nm, and the initial cell concentration was adjusted to OD 600nm 0.01 for antimicrobial assay.

Preparation of Chlorophyllin
Chlorophyllin was prepared from spinach, as described previously [9]. In brief, 500 g of frozen spinach was thawed, mixed with 5 g CaCO 3 to avoid phaeophytin formation, and pressed through a cotton tissue to remove water. In the darkness, the leaves were extracted in 500 mL methanol for 12 h at 4 • C. After filtration, 300 mL petroleum benzene was added, forming a hydrophobic phase on top of the methanolic phase. Subsequent addition of small aliquots of saline water (5% NaCl) forced the chlorophyll into the benzene layer. After the complete transfer of chlorophyll into the benzene phase, phases were separated using a separating funnel. A defined small sample volume was withdrawn, the benzene was evaporated, and the residual chlorophyll was dissolved in a specified volume of methanol. The chlorophyll concentration in the benzene phase was determined according to Lichtenthaler and Wellburn (1983) [44]. Finally, 250 mL benzene-chlorophyll was extracted using 15 mL of 100 mM methanol-sodium hydroxide solution. This treatment cleaves the phytol tail of chlorophyll and transforms it into water-soluble chlorophyllin, which moves into the methanol-sodium hydroxide phase. The volume of the resulting chlorophyllin in the methanol-sodium hydroxide phase was determined, and its concentration was calculated accordingly. All incubations were carried out in the darkness, with sample handling under low light conditions.

Determination of Minimum Bactericidal Effects of PEI and Chlorophyllin Combinations
The minimum bactericidal concentration (MBC) of PEI (25 kDa, branched, cat no. #408727, Sigma-Aldrich, St. Louis, MO, U.S.A.) and chlorophyllin combination was determined as described [9]. In brief, cells were incubated in 96-well plates for defined times with different concentrations of PEI and chlorophyllin. The initial cell concentration was adjusted to an optical density (OD 600nm ) of 0.01. For each experiment, two plates were prepared. One plate was exposed to light at an intensity of 12.7 mW cm −2 , and the other plate was wrapped with aluminum foil, which was used as a dark control. A spectrum of the light source is shown elsewhere [9]. At defined time points of incubation, 3 µL from each sample were plated on LB agar and incubated overnight at 37 • C. Subsequently, colony formation was determined, where no growth was classified as 100% effective, strongly reduced growth as 66%, reduced growth as 33%, and complete growth as 0% inhibitory effects, if not otherwise indicated. The number of replications is indicated in the corre-sponding figures. The mean values of individual bactericidal values from independent experiments were calculated and displayed.

Statistical Analysis
The data were analyzed by Combenefit software (version 2.021) https://www.cruk. cam.ac.uk/research-groups/jodrell-group/combenefit (accessed on 23 July 2022). The dose-response matrix shows the eradication percentage (the darker the color, the higher the eradication effect). Loewe synergy model was used to determination of the synergism and the antagonism between the tested combinations. In the Loewe synergy and antagonism matrix, N represents the number of biological replicates.

Determination of PEI/Chlorophyllin Effects on Cell Density
Experiments were performed with overnight (16 ± 2 h) grown cultures of E. coli DH5α (OD 600nm 2.92). Cell density gradients were prepared in 96-well plates in a horizontal direction (left to right of the plate), with two-fold serial dilution of cells in LB medium from one column to the next. Subsequently, two-fold serial dilution of PEI was in the vertical direction (top to bottom of the plate). The last column represents growth control. Defined chlorophyllin concentrations (1.25, 10, and 30 µg/mL) were added separately. After incubation for 24 h under light (12.7 mW cm −2 ) or darkness, 3 µL from each well were plated on LB agar and incubated overnight at 37 • C as described in Section 4.3. PEIchlorophyllin combinations were classified as having no eradication effect if any colony was observed on the corresponding inoculated spot, while the complete absence of cells was classified as an effective eradication concentration.

Conclusions
Combinations of chlorophyllin and PEI exhibit strong antibacterial effects, above all under illumination but to a lower extent also in darkness. The results of this study are in context with observations of other research groups and suggest that the potential for synergistic effects of cell wall destabilizing substances (e.g., polyamines) and photosensitizers may exist. The chemical structures can be modified to improve antimicrobial activity further. Furthermore, the determination of antimicrobial properties and cell toxicity of different PEIs or PEI-nanoparticles are of utter interest. At least under light conditions, antimicrobial effects of chlorophyllin are most likely caused by induction of excessive oxidative stress, which makes the formation of resistant strains very unlikely.